Incremental vs Discontinuous Tasks
A taxonomy of intellectual tasks the 2023 Contrary Research report borrows from Microsoft Research's "Sparks of AGI" paper: incremental tasks are solved sequentially or gradually, with progress made one word or sentence at a time (summaries, factual Q&A, composing within a known form) — a shape that suits autoregressive next-token prediction. Discontinuous tasks require a creative leap or a new framing of the problem (jokes, novel scientific hypotheses, new styles) — a shape the report states 2023-era models were bad at. The taxonomy's use is routing: it predicts which delegations to a model will disappoint before you try them, and it explains the layman's whiplash ("computers aren't so smart after all") as a category error — asking a discontinuous question of an incremental engine.
The report's second-brain application: making incremental connections across a large store of existing notes is squarely in the incremental class, so an LLM over your knowledge base plays to the model's strength even when open-ended creativity does not.
Claims
- Intellectual tasks divide into incremental (gradual, sequential progress) and discontinuous (requiring a creative leap or reframing); the report states GPT-4-class models are strong at the former and weak at the latter. (observation — the source's 2023 characterization of model capability, citing "Sparks of AGI"; check-worthy and dated)
- Route LLM delegation by task shape: point the model at connection-making and synthesis over stored material rather than at creative leaps. (best_practice — context: the report's advice for 2023-era autoregressive models; the boundary between the classes is itself provisional as models change)
Related
- Model-Tier Routing — the sibling routing axis: this taxonomy routes by task shape, tier routing by task depth/cost.
- AI Second Brain — the application: connection-making over the store is incremental, hence LLM-suitable.
- Dynamic Retrieval — what the incremental engine does over the store once retrieval surfaces the material.
- Distillate: From Notetaking to Neuralink (Contrary Research)